The Cooplexity model at Café Rouge

Café Rouge Oxford

Some months ago I went to the Café Rouge Restaurant in Oxford and we had one of the most interesting service encounters ever.

I was having dinner with my wife. This was not the first time we’d visited it and on previous occasions we had tried a very nice wine that was recommended by one of the waiters. We couldn’t remember the name so we tried a different one. In the middle of the meal, when we had consumed half the bottle we recognised the waiter from our last visit and asked for the same bottle of wine. He approached us from his area (waiters have a designated area of responsibility) and told us that yes, he remembered us and the wine was a Malbec, a kind of wine originally from the south west of France (Burdeos and Cahors). In fact, he remembered even where we had been sitting at the time. I don’t have a very good memory for names and places so when these things happen they impress me a lot.

After a while our waiter came and asked us if everything was ok or whether we needed anything. The question was obvious because we had just been talking with this other waiter and he probably noticed. We explained the situation and how we’d mistakenly chosen the wrong wine because we hadn’t been able to recall the name. What happened after was not so straightforward. He went to ask something to our known waiter and immediately after he talked to another person that we couldn’t identify at that time. Then he came and offered to change the bottle for the Malbec, but we declined because the current wine was also good and we had already drunk half of it. But he insisted so much that we finally accepted. In a moment we were drinking our preferred wine.

The level of satisfaction was very high and the associated feelings extremely gratifying. The situation was so unusual that made me think deeper. My first analysis was from the waiter’s perspective. I made an association with the situation and the John Poindexter(1) Model (Zamora Enciso, 2011 pag. 68). For those who are not familiar with it, his model, based on the Knowledge Pyramid (DIKW) from Russell Ackoff (Ackoff, 1989) distinguishes between:

  • Data. This is gross data. In this case the bottle of wine we had.
  • Information. Data in context. Understanding the situation. We wanted a different wine.
  • Knowledge. Understanding what the information means. We had a kind of feeling of frustration. That is why we asked the waiter who had served us before.
  • Options. We decided what to do according to the constraints. Clearly in this case change the bottle or pass.
  • Action. Execution. Asking for permission from the boss who was the person we initially didn’t identify and change it.
  • Finally we had an interaction that produced a great satisfaction.

But then I began to consider a couple of simple questions from the business perspective. How was it possible to have workers so involved? What was the underlying policy? The first thing we have to notice is the waiter’s proactivity, which means a lot. The Oxford Dictionaries define proactive as an adjective “(of a person or action) creating or controlling a situation rather than just responding to it after it has happened”. Proactivity is central to the Cooplexity model (Zamora Enciso, 2011), a model of collaboration in complexity for management in times of uncertainty and change. Complexity is related to emergent behavioural patterns and human interrelations are clearly complex because of our capacity to make different decisions. According to the model, “proactivity oriented to results” and “proactivity oriented to relations” are key factors of the first level or knowledge level. In both cases we recognize the activities of the model (data gathering, decision making, control of the objective first and interaction, interchange, relation after) in relation to the waiter’s actions.

At this point it is worth thinking about their objectives? For sure it is not profitability because it was possible to avoid the expense. Instead it is easy to assume the option of emotional satisfaction and therefore loyalty (we have repeatedly gone back since then). When I remember the situation I can still visualise a kind of emotional connection. It couldn’t just be acting. I know what is called Emotional Labour, a term coined by the sociologist Arlie Russell Hochschild as “the management of feeling to create a publicly observable facial and bodily display” (Hochschild 1983). His work, explained from the perspective of theatre where the customer is the audience and the service provider the actor, proposes that managing emotions is one way for employees to achieve organizational goals. Thus, there are two processes involved, surface acting (managing observable expressions) and deep acting (managing feelings).

Not enough. It cannot be so simple. When the waiter that attended our table went to ask his colleague, it seems he was checking the wine we wanted. This approach must only be done based on a trusting relationship. Any other situation of mistrust would result in a lack of action. We can clearly find again a parallelism with the second level of the Cooplexity model, which is related to cohesion. In this case the awareness of a common project, and the two factors “group integration” (cooperation, implication) and specifically “trust generation” are present.

Finally, in order to ensure the emergence of self-coordination (third level of the model), it must be a policy oriented to people, supporting initiatives, encouraging ideas, creating a decentralized environment, giving them a certain level of autonomy, demanding results instead of procedures, asking for responsibility and not just the standard expected performance, etc. When the situation is complex, an executive cannot manage the whole business, cannot be everywhere, cannot make all the decisions and cannot be aware of everything. The identification of “local” opportunities and risks depends on the people who are in contact with a situation which demands the last two factors of the model, “equal relationship” (mutual consideration, respect) and “criterion of action” (definition of a criterion). The definition of a criterion is not only a problem of defining the objective (already done) but also the way it can be reached. Therefore acting and empathizing are two key dimensions of this criterion.

That is when I can clearly apply the concept of high performance team to all of them as “a small number of interdependent persons that are spontaneously and naturally coordinated, with the motive of a common project, thanks to a feeling of membership resulting from a determined level of cohesion, making decisions based on shared knowledge” (Zamora Enciso, 2011 pag. 7).

But if I include us having dinner then we were a system considered as “a series of parts that interact with each other to work as a whole. Nevertheless, a system is more than the sum of its parts; it is the product of its interactions (Kauffman, 1980).

And it works. I can assure you!


(1) John Poindexter was director of the Information Awareness Office (IAO), an official body depending of the Defence Advanced Research Projects Agency (DARPA), responsible of the development of projects like ARPANET, the first information transmission network by packets, predecessor of the present Internet.



Ackoff, R. L. (1989). From Data to Wisdom. Journal of Applied Systems Analysis 16, 3-9.

Hochschild, A. R. (1983). The managed heart: Commercialization of human feeling. Berkeley, CA: University of California Press.

Kauffman, D. L. (1980). Systems One: An Introduction to System Thinking. Mineapolis, MN: Future Systems.

Zamora Enciso, R. (2011). Cooplexity: A model of collaboration in complexity for management in times of uncertainty and change.